Massively Parallel Grid Parameter Search for scRNA-seq Analysis

Easy & Quick : Docker services are using hyperopt in parallel on top of Mongodb

A wrapper for searching over Neural network architectures, built on top of MXNet framework, in particular gluon.

Hyperparameter Optimization Using Sci-kit Optimization

Additional Callbacks for Weights & Biases to monitor your models even better 🔎

Code written in Python 2.7 for experiment on SVM with RBF kernel on "Robust and Efficient Kernel Hyperparameter Paths with Guarantees".

MLCloudSearch presents tooling built to facilitate the use of Scikit-Learn and Tensorflow Keras Models. The name MLCloudSearch was inspired by the Machine Learning technique of searching for the best parameters to use in fitting a model, commonly referred to as Hyper-Parameter Search. Training can be done locally or on the Cloud.

I have trained two different CNN models for binary image classification to see which architecture has better accuracy, takes less time in training, how hyperparamters affect training and how many epochs do each of them need. I achieved 96% accuracy on the best model.

Example of Neu.ro integration with NNI for hyperparameter tuning

A lightweight custom automl library.

Hyperparameter selection on machine learning models using Particle Swarm Optimization

Hyperparameter search algorithm

Machine Learning Term project of IOG5016 and IOG5018

Tree-of-Parzen-estimators hyperparameter optimization

PyTorch template to facilitate scientific research using deep learning

Project presented in the parallel and distributed computing course of the Graduate Program in Computer Science at CEFET-RJ (2018.3)

A specialized repository for fitting the Vikhlinin model to galaxy cluster density profiles.

Easy parameter specification using dataclasses made using attr library.

Использование MLflow для трекинга экспериментов PyTorch и Sklearn

hyperparameter search for machine learning